Resurrection of DNA Function In Vivo from an Extinct

Genome
Andrew J. Pask1,2, Richard R. Behringer1*, Marilyn B. Renfree2
1 Department of Molecular Genetics, University of Texas M. D. Anderson Cancer Center, Houston, Texas, United States of America, 2 Department of Zoology, University of
Melbourne, Melbourne, Victoria, Australia

Abstract
There is a burgeoning repository of information available from ancient DNA that can be used to understand how genomes
have evolved and to determine the genetic features that defined a particular species. To assess the functional consequences
of changes to a genome, a variety of methods are needed to examine extinct DNA function. We isolated a transcriptional
enhancer element from the genome of an extinct marsupial, the Tasmanian tiger (Thylacinus cynocephalus or thylacine),
obtained from 100 year-old ethanol-fixed tissues from museum collections. We then examined the function of the enhancer
in vivo. Using a transgenic approach, it was possible to resurrect DNA function in transgenic mice. The results demonstrate
that the thylacine Col2A1 enhancer directed chondrocyte-specific expression in this extinct mammalian species in the same
way as its orthologue does in mice. While other studies have examined extinct coding DNA function in vitro, this is the first
example of the restoration of extinct non-coding DNA and examination of its function in vivo. Our method using
transgenesis can be used to explore the function of regulatory and protein-coding sequences obtained from any extinct
species in an in vivo model system, providing important insights into gene evolution and diversity.

Citation: Pask AJ, Behringer RR, Renfree MB (2008) Resurrection of DNA Function In Vivo from an Extinct Genome. PLoS ONE 3(5): e2240. doi:10.1371/
journal.pone.0002240
Editor: Erik I. Svensson, Lund University, Sweden
Received January 4, 2008; Accepted April 16, 2008; Published May 21, 2008
Copyright: ß 2008 Pask et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Supported by NSF grant IBN 0220458 and the Ben F. Love Endowment to RRB, NHMRC CJ Martin and RD Wright Research Fellowships to AJP, and an
ARC Federation Fellowship to MBR. DNA sequencing and veterinary animal care was supported by the NIH Cancer Center Support Grant CA16672.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: rrb@mdanderson.org

Introduction genomic DNA. Recent advances in sequencing technologies, such

Extant species represent less than 1% of the genetic diversity
that has existed in the animal kingdom [1]. Extinction rates are
increasing at an alarming rate, especially of mammals [2,3]. Many
efforts, such as that of the Frozen Zoo (San Diego Zoological
Society Conservation and Research for Endangered Species) are
working to cryo-archive cell and tissue resources from a diverse
range of threatened species, to protect their genetic information.
However, for those species that have already become extinct,
access to their genetic biodiversity may not be completely lost.
The Tasmanian tiger or thylacine (Thylacinus cynocephalus) was a
large, carnivorous Australian marsupial. Often described as the
most striking example of convergent evolution in the mammalian
lineage, the marsupial thylacine was morphologically almost
indistinguishable from the eutherian canids, apart from the
presence of a pouch where its young developed (Figure 1) [4].
Thylacines were hunted to extinction in the wild in the early 1900s
and the last known animal died in captivity in the Hobart Zoo in
1936 [4]. Fortunately, some thylacine pouch young and adult
tissues were preserved in alcohol in several museum collections
around the world (Figure 1b).
With improved techniques for the isolation of ancient DNA, it is
now possible to access the genomes of extinct species [5]. There
have been several recent papers published examining the genomes
from a diverse range of extinct species from plants and bacteria to
mammoths and Neanderthals [2,5–7]. While most studies to date
have examined the mitochondrial DNA of these species for
phylogenetic purposes, the focus is now shifting to the analysis of
as direct high-throughput parallel pyrosequencing, are expected to
expand the ancient DNA data bank exponentially over the coming
years [5]. Most studies have used the sequence data to examine
divergence times and population structure of extinct species.
However, more recently this information is being used to examine
how the function of these genes may have evolved. The function of
the Melanocortin 1 Receptor (MC1R) gene has recently been
investigated from the extinct mammoth and Neanderthal [1,7].
Isolated MC1R sequences from ancient DNA samples were cloned
and transfected into cell lines to examine the function of the
receptor in activation assays in vitro. The results of these
experiments suggest that variation in skin and hair pigmentation
may have occurred within these species [1,7] and describes a
technical platform for examining extinct protein function in vitro.
While many adaptive changes throughout evolution have been
ascribed to changes within the proteins themselves [8,9], the high
conservation of protein coding regions between mammalian
genomes suggests that changes in the open reading frames are
unlikely to be the primary cause of the vast differences observed in
both form and function [10–12]. With the recent release of the
ENCODE pilot study [13] it is clear that the majority of the non-
coding mammalian genome is transcribed. It appears that rather
than changes in the genes themselves, it is subtle differences in
their non-coding regulatory elements that control their spatial,
temporal and quantitative expression that underpins the variation
in the animal kingdom [10–12]. Therefore, the non-coding regions
of extinct genomes are likely to hold the most important genetic

PLoS ONE | www.plosone.org 1 May 2008 | Volume 3 | Issue 5 | e2240


Figure 1. The thylacine, Thylacinus cynocephalus. (a) Young male
thylacine in Hobart Zoo in 1928, photograph (Q4437). (b) One of the
preserved pouch young specimens (head length 34 mm) from which
DNA was extracted, from the Museum Victoria collection. (c-f) The skull
of the thylacine (c,e) compared with that of the domestic dog Canis
canis (d,f). The morphology of the head shows remarkable convergent
evolution. However, there are some differences: in marsupials, the
lacrymal extends outside the orbit and the angle of the dentary is
medially inflected (c). The thylacine palatine has the vacuities
characteristic of marsupial skulls (e). The teeth also show striking
convergent evolution but the muzzle of the thylacine is quite narrow
compared to that of the dog (e,f). Scale bar = 5cm.
doi:10.1371/journal.pone.0002240.g001
information that defined a species. We have therefore developed a
technique for examining the function of non-coding regions from
extinct genomes in vivo. To gain a greater insight into the function
of these elements we isolated a regulatory sequence from the
genome of the extinct thylacine using transgenesis and examined
its function in mice.
Results
Extinct DNA Isolation and Fragment Characterisation
To resurrect the function of non-coding DNA from an extinct
mammal, we isolated genomic DNA using the technique described
by Pääbo [14]. DNA was obtained from four 100-year-old
specimens: three alcohol-fixed pouch young (one shown in
Figure 1b) and one dried adult pelt, obtained from Museum
Victoria (Melbourne, Victoria, Australia). The isolated DNA was
fragmented as expected [6], and ranged from 300–500 bp. We
obtained 200 ng to 1 mg from each sample. We chose to isolate the
well-characterised transcriptional enhancer element of the
proa1(II) collagen (Col2a1) gene [15–17]. This element was chosen
because it is relatively conserved among mammals and directs
chondrocyte-specific expression in the mouse [15,17]. Primers
were designed to span the core enhancer from regions conserved
PLoS ONE | www.plosone.org
In Vivo DNA Resurrection
across all described mammalian species using a Clustal-W
alignment. Primers were engineered with Spe1 (forward primer)
and Xba1 (reverse primer) restriction sites to allow for multi-
merisation of the resulting product (see Materials and Methods).
We performed PCR amplification from each DNA sample
independently to reduce the risk of cross contamination, and to
ensure the sequence was identical across all individuals, confirming
it was thylacine in origin. This approach was also used to ensure
sequence obtained was not harbouring any genetic changes caused
by spontaneous hydrolysis or oxidation of the preserved DNA [6].
The resulting 264-bp PCR product from each sample, encoding
the enhancer was independently subcloned and sequenced. The
sequence obtained from each of the four independent thylacine
samples was identical and a BLAST search confirmed it was
different from all those available in the database and from the most
likely sources of contamination in the lab (human and mouse)
(Figure 2a). Phylogenetic analyses of the thylacine sequence with
that of eutherian and marsupial species confirmed its identity as
the thylacine Col2a1 enhancer orthologue (Figure 2b), as it was
most closely related to the tammar wallaby (Macropus eugenii, an
extant marsupial species) than to human, mouse or rat.
Figure 2. Analyses of the thylacine Col2a1 enhancer element. (a)
Sequence alignment of the thylacine PCR cloned Col2a1 enhancer.
Identical alignment between species is shown by black boxes. The
thylacine sequence was most similar to, but distinct from that of
another marsupial, the tammar wallaby, Macropus eugenii. The black
box outlines the minimal 18-bp element. Inverted repeats (arrows)
surrounding the minimal element are shown with complementary
nucleotides in red. The outer repeat is highly conserved between all
mammals, but the inner repeat differs by 2 nucleotides in the thylacine.
(b) Phylogenetic analysis of the thylacine Col2a1 enhancer. The
thylacine sequence groups with the tammar wallaby, both of which
are more similar to the human sequence than to mouse or rat. Numbers
indicate bootstrap values based on 100 replicates.
doi:10.1371/journal.pone.0002240.g002
2 May 2008 | Volume 3 | Issue 5 | e2240
 In Vivo DNA Resurrection

Transgenesis and Functional Analyses

The strength of the Col2a1 regulatory element can be enhanced
by using multiple copies of this region upstream of a reporter gene
[16]. Four copies of the thylacine sequence were multimerised and
ligated to the human b-globin basal promoter fused to lacZ and
followed by a polyadenylation signal. The 4.75 kb TcyCol2a1-lacZ-
pA construct (Figure 3a) was purified from the vector and
microinjected into the pronuclei of mouse zygotes [18]. LacZ
expression was examined in founder mice at 14.5 days post coitum,
by staining with X-gal [18] (5-bromo-4-chloro-3-indolyl-b-D-
galactopyranoside) (Figure 3b–e).
Whole mount staining was observed in all 4 founder fetuses
created in a pattern indicative of developing cartilage (Figure 3b–
e). This pattern was identical to that of the endogenous mouse
Col2a1 gene and mouse enhancer transgenes [15–17]. Histological
cross sections of the developing limb confirmed that lacZ
expression, directed by the thylacine enhancer element, was
restricted to the developing chondrocytes (Figure 3g). Non-
transgenic littermates were used as negative controls (Figure 3f)
and showed no staining.

Discussion
A non-coding DNA fragment from an extinct mammal was able
to drive expression of a reporter gene construct in a developing
fetus. While the intensity of the reporter gene expression seen in
the transgenic fetuses varied, the sites of expression did not.
Variations in transgene expression levels are common, and can be
affected by the site of integration and by the effect of epigenetic
silencing mechanisms on the transgene [18]. Despite the level of
variation, the expression of the thylacine Col2A1 enhancer
recapitulated that of the endogenous mouse Col2a1 gene [15–17]
and was restricted to the developing chondrocytes. This confirms
that the thylacine Col2a1 gene had a conserved developmental role
in cartilage formation, and that its promoter directed expression in
chondrocytes in this extinct marsupial mammal. It also suggests
that the associated transcription factors and the minimal enhancer
element required for inducing expression from this promoter have
remained sufficiently conserved between mice and marsupials to
direct expression [19,20] despite approximately 148 million years
of divergent evolution between these species [21–23].
The minimal 18-bp element [15] required for chondrocyte
specific expression is only 17-bp in the thylacine and differs at 4
nucleotides positions compared with the mouse. Furthermore, the
sequences of the inverted repeats thought be to essential for the
function of this region that flank the minimal element are
completely complementary in mouse, rat and human, [15].
However, in the thylacine, the outer repeat is highly conserved Figure 3. From extinction to gene expression. Functional analysis
but the inner repeat differs by 2 nucleotides and is not of the thylacine non-coding DNA fragment. (a) Diagram of transgene
construct. 4 copies of a 264-bp fragment containing the Thylacine
complimentary. Despite these differences, the enhancer element Col2a1 enhancer (TcyCol2a1) region was ligated to the human b-globin
was able to function in the developing mouse, suggesting that these minimal promoter (black box) and ligated to lacZpA. (b–e) X-gal stained
differences are not critical for the function of this element. 14.5 dpc TcyCol2a1-lacZpA transgenic mouse embryo showing varying
Therefore, the examination of extinct DNA can provide important levels of reporter gene expression within the developing cartilage
information about gene function in extant species. Taken together, (blue). (f) Non-transgenic littermate, negative control fetus. (g) Top
panel; Magnified image of forelimb from fetus in (b) black line indicates
the results of this study confirm an ancestral role for the thylacine
the plane of section shown in (g) bottom panel. Bottom panel;
Col2A1 gene in cartilage formation and demonstrate that it is Histological section of transgenic forelimb digit, showing lacZ-
possible to examine non-coding DNA function from extinct expressing chondrogenic tissue (blue) counterstained with eosin (pink).
animals in vivo in the mouse. doi:10.1371/journal.pone.0002240.g003
These observations advance previous studies that have so for
only examined extinct protein function and only in in vitro systems
[1,7]. However, both systems (in vitro cell cultures and in vivo whole receptor signalling strength seen in the mammoth and Neander-
organism models) rely on the availability and compatibility of thal MC1R gene analyses in vitro [1,7] did not occur or affect gene
cofactors within the host cell or embryo in order to examine gene function in their native organisms. The native ligand in the
or regulatory sequence function. It is possible that the variations in mammoth and Neanderthal may have varied to the one used in

PLoS ONE | www.plosone.org 3 May 2008 | Volume 3 | Issue 5 | e2240


culture, as could the downstream activation strength of the
signalling pathway. Similarly it is possible that the thylacine
Col2A1 enhancer had a different pattern of expression in the
developing thylacine embryo to that observed in the mouse, due to
the different availability of cofactors required for its activation.
The precise function of an extinct gene or regulatory element is
impossible to determine without examining every part of the
pathway in one system. Results from extinct DNA analyses
therefore should be cautiously interpreted.
Thus extinct genomes can be examined by the in vivo method
described here, in conjunction with in vitro techniques, to enable
functional analyses of the genes and regulatory elements. This
could provide important insights into genome evolution by
defining functional features unique to each species. We have
demonstrated that non-coding genetic information from an extinct
species can be resurrected in vivo and in doing so, we have restored
to life the genetic potential of a fragment of this extinct
mammalian genome.
Materials and Methods
Tissues, DNA isolation, amplification, and transgene
construction and analyses
Tissues were obtained from three 100-year-old thylacine pouch
young specimens fixed in ethanol and one dried 100-year-old adult
skin from Museum Victoria, Melbourne. Tissue samples were
stored in sterile cryo-vials and transported to the University of
Texas. DNA was extracted according to the method described by
Pääbo [14] with a control extraction performed on an empty vial.
All processing was done in a UV-sterilised bio-containment hood
to prevent any contaminating DNA from entering the samples.
DNA samples were kept separate and used as separate templates in
subsequent PCR reactions. Negative controls (blank-DNA extrac-
tion, and no template) were included in each PCR reaction to
ensure no contamination of samples. Primers were engineered
with Spe1 (forward primer) and Xba1 (reverse primer) restriction
sites to allow for multimerisation (forward 59 NNNACTAGTG-
CATTGGGAGATTGGCAGCGAT 39; reverse 59 NNNTCTA-
GAGCTACCTCTTTCGGGGAACTG 39). The resulting 264-
bp PCR product was subcloned and sequenced in both directions
for at least three clones from each tissue sample. The sequence was
compared to that from possible sources of contamination within
References
1. Römpler H, Rohland N, Lalueza-Fox C, Willerslev E, Kuznetsova T, et al.
(2006) Nuclear gene indicates coat-color polymorphism in mammoths. Science
313: 62.
2. Novacek MJ, Cleland EE (2001) The current biodiversity extinction event:
scenarios for mitigation and recovery. Proc Natl Acad Sci USA 98: 5466–70.
3. Hilton-Taylor C (compiler) (2000) 2000 IUCN Red List of Threatened Species, IUCN,
Gland, Switzerland and Cambridge, UK.
4. Paddle R (2000) The Last Tasmanian Tiger. Cambridge: Cambridge University
Press.
5. Green RE, Krause J, Ptak SE, Briggs AW, Ronan MT, et al. (2006) Analysis of
one million base pairs of Neanderthal DNA. Nature 444: 330–336.
6. Willerslev E, Cooper A (2005) Ancient DNA. Proc R Soc B 272: 3–16.
7. Lalueza-Fox C, Römpler H, Caramelli D, Stäubert C, Catalano G, et al. (2007)
A melanocortin 1 receptor allele suggests varying pigmentation among
Neanderthals. Science 318: 1453–5.
8. Coyne JA, Hoekstra HE (2007) Evolution of protein expression: new genes for a
new diet. Curr Biol 17: R1014–6.
9. Hoekstra HE, Coyne JA (2007) The locus of evolution: evo devo and the genetics
of adaptation. Evolution Int J Org Evolution. 61: 995–1016. Rev.
10. King MC, Wilson AC (1975) Evolution at two levels in humans and
chimpanzees. Science 188: 107–16.
11. Carroll SB (1995) Homeotic genes and the evolution of arthropods and
chordates. Nature 376: 479–85.
12. Prud’homme B, Gompel N, Carroll SB (2007) Emerging principles of regulatory
evolution. Proc Natl Acad Sci USA. 104 Suppl 1: 8605–12.
PLoS ONE | www.plosone.org
In Vivo DNA Resurrection
the lab using genome databases (human, mouse and tammar
wallaby) and was shown to have consistent differences with each
species clearly indicating it was derived from a unique source. The
fragment was also subjected to phylogenetic analysis to confirm its
marsupial origin. Phylogentic relationship of the thylacine PCR
fragment was compared to the orthologous region from tammar
wallaby, human, mouse and rat using the PHYLIP 3.63 program
(University of Washington; http://evolution.genetics.washington.
edu/phylip/doc/main.html) using maximum-likelihood, maxi-
mum parsimony and neighbor-joining analysis with 1000
replicates. The resulting data was viewed with TREE-view 1.6.6
(http://taxonomy.zoology.gla.ac.uk/rod/treeview.html).
Once sequence identity was confirmed, the clone was digested
with Xba1 and Spe1 and incubated with T4 ligase at 16oC for
2 hours. After ligation, the reaction was digested with Xba1 and
Spe1 such that only products multimerised head to tail (ablating the
restriction sites) would remain intact. Ligation products corre-
sponding in size to 4 mutimerised copies (1056 bp) were purified.
The multimer was ligated to a human b-globin basal promoter
fused to lacZ followed by a polyadenylation signal. Embryos were
stained for LacZ activity [17]. The human b-globin basal
promoter-lacZ-pA reporter construct on its own does not have
tissue-specific activity in transgenic mice [24]. For histological
analyses, X-gal stained embryos were postfixed in 4% parafor-
maldehyde, embedded in paraffin, sectioned and counterstained
with eosin [17].
Acknowledgments
We thank Dr. Joan Dixon and Lina Frigo and the Museum Victoria for
allowing us to sample rare thylacine material. We thank George Adebayo
for pronuclear injections, Ying Wang for histology. We also thank Dr.
Deanne Whitworth, Dr. Chris Cretekos and Richard Moyle for assistance
and Professor Roger Short for helpful comments. We especially thank the
Tasmanian Museum and Art Gallery for the kind permission to reproduce
photograph (Q4437) by Benjamin A. Sheppard, David Paul for the
photograph of the thylacine pouch young and Associate Professor Geoffrey
Shaw for the photographs of the dog and thylacine skulls.
Author Contributions
Conceived and designed the experiments: RB AP MR. Performed the
experiments: RB AP. Analyzed the data: RB AP. Contributed reagents/
materials/analysis tools: MR. Wrote the paper: RB AP MR.
13. The ENCODE Project Consortium (2007) Identification and analysis of
functional elements in 1% of the human genome by the ENCODE pilot
project. Nature 447: 799–816.
14. Pääbo S (1989) Ancient DNA: extraction, characterization, molecular cloning,
and enzymatic amplification. Proc Natl Acad Sci U S A 86: 1939–43.
15. Lefebvre V, Zhou G, Mukhopadhyay K, Smith CN, Zhang Z, et al. (1996) An
18-base-pair sequence in the mouse proalpha1(II) collagen gene is sufficient for
expression in cartilage and binds nuclear proteins that are selectively expressed
in chondrocytes. Mol Cell Biol 16: 4512–23.
16. Zhou G, Garofalo S, Mukhopadhyay K, Lefebvre V, Smith CN, et al. (1995)
A 182 bp fragment of the mouse pro alpha 1(II) collagen gene is sufficient
to direct chondrocyte expression in transgenic mice. J Cell Sci 108:
3677–84.
17. Leung KK, Ng LJ, Ho KK, Tam PP, Cheah KS (1998) Different cis-regulatory
DNA elements mediate developmental stage- and tissue-specific expression of
the human COL2A1 gene in transgenic mice. J Cell Biol 141: 1291–300.
18. Nagy A, Gertsenstein M, Vintersten K, Behringer R, eds (2002) Manipulating the
Mouse Embryo, A Laboratory Manual, 3rd Edition. New York: Cold Spring Harbor
Press.
19. Lefebvre V, Huang W, Harley VR, Goodfellow PN, de Crombrugghe B (1997)
SOX9 is a potent activator of the chondrocyte-specific enhancer of the pro
alpha1(II) collagen gene. Mol Cell Biol 17: 2336–46.
20. Lefebvre V, Li P, de Crombrugghe B (1998) A new long form of Sox5 (L-Sox5),
Sox6 and Sox9 are coexpressed in chondrogenesis and cooperatively activate the
type II collagen gene. EMBO J 17: 5718–33.
4 May 2008 | Volume 3 | Issue 5 | e2240

21. Bininda-Emonds OR, Cardillo M, Jones KE, MacPhee RD, Beck RM, et al.
(2007) The delayed rise of present-day mammals. Nature 446: 507–12.
22. Luo ZX, Ji Q, Wible JR, Yuan CX (2003) An Early Cretaceous tribosphenic
mammal and metatherian evolution. Science 302: 1934–40.
PLoS ONE | www.plosone.org
In Vivo DNA Resurrection
23. Wible JR, Rougier GW, Novacek MJ, Asher RJ (2007) Cretaceous eutherians
and Laurasian origin for placental mammals near the K/T boundary. Nature
447: 1003–6.
24. Yee SP, Rigby PW (1993) The regulation of myogenin gene expression during
the embryonic development of the mouse. Genes Dev 7: 1277–89.
5 May 2008 | Volume 3 | Issue 5 | e2240